ABSTRACT Genetically encoded drug biosensors hold great promise as an enabling pharmacokinetic tool for understanding adverse opioid side effects and finding effective ways to dissociate them from therapeutic effects. However, creating such biosensors for any given drug target with desired sensitivity, selectivity, and kinetics is an unsolved problem. The long-term goal is to develop and use drug biosensors to elucidate the mechanisms underlying opioid side effects and guide the design of new analgesic therapeutics. The overall objective of this application is to establish an efficient method for creating fluorescent biosensors for real-time single-cell and subcellular measurement of opioids. The PI?s laboratory recently developed a COMBINES-CID method for de novo engineering of chemically induced dimerization (CID) systems, in which two proteins dimerize only in the presence of a small-molecule ligand. It is hypothesized that opioid-induced dimerization systems can be generated by COMBINES-CID and then converted to biosensors by coupling them to a fluorescence readout for real-time pharmacokinetic imaging. This hypothesis will be tested by pursuing two specific aims: 1) Generate CID systems with high sensitivity and fast kinetics for fentanyl, the most widely used synthetic opioid, to provide a proof-of-principle; and 2) Engineer CID systems into fluorescent biosensors and validate their performance in neurons differentiated from human induced pluripotent stem cells (iPSCs). Under the first aim, the vastly diverse combinatorial libraries of nanobodies, monobodies, and computationally designed proteins will be screened to obtain CID binders: i) ?anchor binders? that first bind to fentanyl and ii) ?dimerization binders? that only bind to the anchor binder-fentanyl complexes not the unbound anchor binders. Obtained CID systems will be further optimized to improve the sensitivity and kinetics by mutagenesis followed by SMI-seq-enabled quantitative protein-protein interaction screening. For the second aim, CID will be coupled to two fluorescence readouts, fluorescence resonance energy transfer and a circularly permuted fluorescent protein, to create fluorescent biosensors. Finally, obtained biosensors will be genetically encoded in HEK293 and HeLa cells and iPSC-derived neurons to measure drug concentrations in cytoplasm, endoplasmic reticulum, and Golgi apparatus. The proposed project is innovative in that it will for the first time provide an efficient, general solution for creating genetically encoded biosensors for a large variety of drugs, which so far is difficult or impossible to achieve with existing methods. It is significant because obtained opioid biosensors will enable high spatiotemporal analysis of drug action to establish causal links with physiological effects in vivo. The new method will also have wide use in many other fields by largely expanding the biosensor toolkit for drug, metabolites, and signaling molecule detection.